MSQA - Korean: leaderboard
MSQA asks short culture-specific questions written by native speakers in their own language, each with one verifiable answer grounded in local sources. This board covers the Korean-language questions. Answers are free-form, judged for containing the gold answer, and scored as the harmonic mean of the correct share and the correct-given-attempted share, so both wrong answers and refusals cost points (ByteDance Seed, Nanjing University and M-A-P, 2026).
Metric: F-score (%; harmonic mean of CO, the share of fully correct answers, and CGA, the correct share of attempted answers, over the 86 Korean-language questions; free-form answers to natively sourced culture-specific short questions asked in their own language, five runs per model, scored by a Gemini-3.1-Pro gold-answer containment judge). Source: arxiv.org. Saturation forecast: Around February 2027. 21 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 73.9 | #94 |
| 2 | Claude Opus 4.7 | 62.8 | #79 |
| 3 | GPT-5.5 | 62.4 | #49 |
| 4 | GPT-5.4 | 59.7 | #123 |
| 5 | Claude Opus 4.6 | 56.4 | #102 |
| 6 | GPT-5.2 (High) | 46.6 | #176 (GPT-5.2) |
| 7 | DeepSeek V4 Pro | 45.8 | #188 |
| 8 | Seed 2.1 Pro | 45.2 | #135 |
| 9 | Gemini 2.5 Flash | 41.2 | #453 |
| 10 | Seed 2.0 Pro (High) | 34 | #211 (Seed 2.0 Pro) |
| 11 | Seed 2.0 Pro (Medium) | 32.6 | #211 (Seed 2.0 Pro) |
| 12 | Seed 2.1 Turbo | 32 | #160 |
| 13 | Qwen 3.5 Plus (Thinking) | 31.7 | #232 (Qwen 3.5 Plus) |
| 14 | GLM-5 | 29.3 | #260 |
| 15 | Seed 2.0 Lite | 28.7 | #289 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=msqa-korean · How It Works · Data refreshed daily, snapshot 2026-11-13.